- From: Sabine Maennel via GitHub <sysbot+gh@w3.org>
- Date: Sat, 30 Oct 2021 19:30:26 +0000
- To: public-dxwg-wg@w3.org
@makxdekkers and @kcoyle thank you both for your answers. @makxdekkers I do understand that datasets differ and I also know that some distribution formats have metadata build in to them (shapefiles for example can come with metadata). I also get that DCAT does not want to restrict the kinds of dataset, but I stiill think it would be great if it would come with an optional hint on how to add metadata to distributions, that don't come with metadata on their own. The reason for this is that datapublishers and probably everybody tends to put their focus on the things that are explicitely mentioned: it is just normal to forget about all the other stuff that is not mentioned, Therefore there are many open datasets today, with tabular data, where the field descriptions are missing. I agree with @kcoyle that `dct:conformsTo` is not enough in that regard, since it does not link to a description, but just states which standard was applied. In my opinion it would be helpful to have the possibilty to add an attribute similar as in DCAT-US: `distribution → describedBy` (see https://resources.data.gov/resources/dcat-us/#distribution-describedBy). With the usage note there it would help to make datapublisher aware that in certain cases it might be needed to add a link to a more structured description of the tabular structure of the distribution. As an optional field it would not be restrictive on datasets in general and would be sort of a minimal solution, adding some hint on what to provide if a distribution has no meta description build into itself. -- GitHub Notification of comment by sabinem Please view or discuss this issue at https://github.com/w3c/dxwg/issues/1418#issuecomment-955581217 using your GitHub account -- Sent via github-notify-ml as configured in https://github.com/w3c/github-notify-ml-config
Received on Saturday, 30 October 2021 19:30:28 UTC